The 9 mo cycle low

Would occur on 5/16/08. I also find interesting the 13.5 month cycle (which is 9 mo * 1.5), which off the 3/14/07 lows would be around first week of June. Also a 4.5 month cycle (9 mo * .5) off of the 11/23/07 low would mark a similar time frame. A nesting of cycle lows is, I've observed, something to be respected. A nesting of highs haven't made much sense to me either.

I did some looking at the 36 week, I think it passed 10-20 degrees ago.
It's a weakest right now, I know it is a Swenlin staple and he got tricked
by it recently.

Would occur on 5/16/08.

I also find interesting the 13.5 month cycle (which is 9 mo * 1.5), which off the 3/14/07 lows would be around first week of June.

Also a 4.5 month cycle (9 mo * .5) off of the 11/23/07 low would mark a similar time frame.

A nesting of cycle lows is, I've observed, something to be respected. A nesting of highs haven't made much sense to me either.

Hey dc, I'd have to run a couple different ways from different multiples of the fundamental cycle,
I was real close to using the 117 week ( 39 * 4 ) before but went with the 92 week for better fit.

This is a first pass fit of the 39 week. And notice the scales, log spx on left, log 39 wk on the
right axis, 39 week is very week right now.

Would occur on 5/16/08.

I also find interesting the 13.5 month cycle (which is 9 mo * 1.5), which off the 3/14/07 lows would be around first week of June.

Also a 4.5 month cycle (9 mo * .5) off of the 11/23/07 low would mark a similar time frame.

A nesting of cycle lows is, I've observed, something to be respected. A nesting of highs haven't made much sense to me either.

If cycles were exact, don't you think everyone and their brother would be making a crapload of money? I have some theories on why cycles may work: 1. biological - ie menstrual cycle type built into the human brain or psychological 2. economical - due to expiration or some other monetary of market mechanics that tend to overlap 3. natural - ie like related to weather patterns and some other physical phenomenon other than biological Of these, I think that economical cycle based on market or monetary mechanics is probably the most dominant, followed by biological being second. I tend to believe that cycles caused by economic/market mechanics are most prevalent in near term (few years) rather than very long term. The biological or natural probably a lot longer correlation going back decades, etc. The 13.5 month cycle that I like is based on tax related selling. People tend to buy and hold for at least a year before dumping so that they can take advantage of reduced taxes. The 9 month cycle.. I think may be more related to options or futures expiration activities. We've had a level of capitulation that may have reduced the impact of economic/market related cycles just due to the disturbance in the system, and if that's the case, then yes I would agree that the normal 9 month low will likely not be useful. But I will also point out that since the volatility in 3/07 where many smart money had bought, prices are now higher. And if they did not capitulate recently (as prices never really went that much lower than the 3/07 low). They have tax based incentive where they can now sell and lock in gains at favorable rates. With economic uncertainty still high (further housing related turmoil almost agreed by everyone), and the fact retail investors are in a mood to buy, will smart money who bought in 3/07 want to dump now and lock in or hold?

Interesting question -- one could probably write a PhD thesis on this topic.

I believe economic/market cycles are a manifestation of agricultural cycles (found in nature), mostly which follow the lunar calendar.

For example:

"In nature, cycles occur because the biological lags encounter biological constraints, such as maximum carrying capacity
of a species' habitat, an increase in a predator species, or other limiting natural phenomena. When the environment
is constant, these life cycles are relatively uniform and vary, depending on the time it takes for individuals within
each species to reach each stage in its life cycle. When taken over whole populations, these life cycles generate
longer population cycles that we recognize by various names, like the cattle cycle. When the environment is not
constant due to drought, disease, or other phenomena, large variations in the regularity of these cycles can occur.
In agricultural commodities, economic phenomena often provide the limits to commodity population inventories.
Cycles in supplies and prices of agricultural commodities occur because of the aggregate effects of interactions
between the limits imposed by nature and the economic environment and biological lags in life histories of
agricultural commodities. At least in this century, and likely for much longer, cycles in agricultural prices and
profits have occurred while cyclical supplies faced a relatively constant or slowly changing demand." (source:http://www.ers.usda.gov/publications/tb1874/tb1874b.pdf)

Given the environmental disruptions caused by global warming, one has to wonder if changes in agricultural cycles will eventually show up in the form of market deviations (cycles), too.

Edited by beta, 27 April 2008 - 11:19 PM.

I posted this one before...

The ST dominant cycles are 6, 10, 20, 38, 76 trading days currently.
The IT dominant cycles are 8, 16, 22-24, 40-42, 90-94 wks currently.

40-42 wk is the 9.25-9.5 month cycle.

90-94 wk probably bottomed.

The next cycles are 8 wk in early May and 16 and 24 wk cycles later in June in my stuff...

I've been going with the very dominant 8 wk cycles lately as seen in the picture, it is just amazingly dominant.

9 mo cycle is not due until later this year apparently.


(btw, there is no reason to wonder about why the cycles are occuring as long as you can measure their intensities and variations within the bounds of the proper information processing, they are like the electromagnetic signals for me or a transmission of an energy state that I tune into to listen the markets)

Edited by arbman, 28 April 2008 - 01:39 AM.

Good to see that I also arrived at 92 week, but 36 instead of 38 days.
I could easily move my period to 37 days but like 38 less.

I do have the 36 day topping and part of the May 13 low but
presently it is one of the weaker cycles and has the larger 72
damping it through May, it wont be a big player. The 20 day
low May 13 is the one.

Arb, that thing looks like it takes some computing power.
I'm working on instantaneous modulated wave output now.

I'm really perplexed about the changes in the longer cycles
and it hit me tonight listening to an historian. Your failed presidents
mark the end, ie something new. Yet the monitization checks are
not indicating something new quite yet.

I posted this one before...

The ST dominant cycles are 6, 10, 20, 38, 76 trading days currently.
The IT dominant cycles are 8, 16, 22-24, 40-42, 90-94 wks currently.

40-42 wk is the 9.25-9.5 month cycle.

90-94 wk probably bottomed.

The next cycles are 8 wk in early May and 16 and 24 wk cycles later in June in my stuff...

I've been going with the very dominant 8 wk cycles lately as seen in the picture, it is just amazingly dominant.

9 mo cycle is not due until later this year apparently.


(btw, there is no reason to wonder about why the cycles are occuring as long as you can measure their intensities and variations within the bounds of the proper information processing, they are like the electromagnetic signals for me or a transmission of an energy state that I tune into to listen the markets)

The other feature of the fourier phase spectrum is while some cycles are fluctuating +/- around an average phase, cyclically, some are in a continous drift around the clock so to speak. The dont really shift or invert, they fluctuate or drift continuously through the view. The other key understand to the output, is the cyclic modulation of the cycle magnitude. All the major cycles that have been there recently ( 8 years ) are dying off.

Use a windowed fourier transform at least and examine the progress of the cycle periods and magnitudes over the past 2-3 cycles in order to estimate the next sample for your forecast, actually it would be better if you extract the cycle shape from the prices too instead of using a Fourier transform...